Prevalence and risk factors for brucellosis in domestic yak Bos grunniens and their herders in a transhumant pastoralist system of Dolpo, Nepal.

Disease caused by Brucella spp. represents the most common bacterial zoonotic infection worldwide. The distribution and public health impact of these infections in Nepal's mountain regions are poorly characterized. This cross sectional study assesses the burden of brucellosis on transhumant pastoralists and their yak in and around Shey Phoksundo National Park, Nepal. Objectives were to: (1) estimate individual animal prevalence of Brucella-seropositive yak, (2) identify herd- and individual-level risk factors associated with Brucella seropositivity in individual yak, and (3) identify herd-level risk factors associated with reported human brucellosis-like symptoms in a household. A case of household symptoms was defined as the reported occurrence within the previous year of at least one of three acute symptoms (chills, fever, night chills) and one of two chronic symptoms (joint pain, swollen joint(s)) in one or both of two individuals interviewed in a household. Two-hundred-ninety-seven yak from 61 herds were sampled, and 61 household questionnaires were completed. Estimated true prevalence was 0.22 (95% CI: 0.17; 0.28). Poisson regression with generalized estimating equations was used to account for repeated measures within a cluster (herd). Yak in herds reporting abortion occurrence within the previous year were 2.3 times more likely to be seropositive than those in herds not reporting abortion (95% CI: 1.2; 4.2, p = 0.01). For every 10 animal increase in herd number, individual animal seropositivity risk increased by 30% (95% CI: 10%; 50%, p = 0.001). Male yak were 0.7 times as likely to be seropositive as female yak (95% CI: 0.5; 0.9, p = 0.01). Three to five year old yak were 2 times more likely to be seropositive than yak <3 years old (95% CI: 1.3; 3.2, p = 0.003), and yak >5 years old were 4.9 times more likely to be seropositive than yak <3 years old (95% CI: 2.9; 8.1, p < 0.001). Risk of reported brucellosis-like symptoms at the household level was 2 (95% CI: 1.1; 3.5, p = 0.02) times greater for households with herds with >1 reactor, and was 3.6 (95% CI: 1.4; 9.2, p = 0.008) times greater for households reporting the practice of raw milk consumption. These results indicate that yak seropositivity for Brucella spp. is widespread in the region, and is associated with reported human disease. This epidemiologic understanding is essential to the identification of public health opportunities at the interface of Himalayan livestock populations and the transhumant pastoralists that depend on them.